All animal experiments were approved by the University of Miami Animal Care and Use Committee according to the National Institutes of Health guidelines and conform to the Guide for the Care and Use of Laboratory Animals. A total of four transgenic mouse lines were used in this study to conditionally regulate c-Myc expression in endothelial cells. All animals were housed on a 12-h light/dark cycle with free access to food and water unless otherwise stated. Endothelial c-Myc knockout mice were generated by crossing c-Myc ^flox/flox (B6.129S6-Myc ^tm2Fwa /Mmax, Strain #032046, Jackson Laboratory, Bar Harbor, ME, USA) and Cdh5(PAC)-CreERT2 (C57BL/6-Tg (Cdh5-cre/ERT2)1Rha), developed by Dr. Ralph Adams at Cancer Research UK (Cancertools.org reference number 151520) mouse lines. Knockout controls consisted of littermates carrying identical floxed genotypes, but lacking Cre (Cre-negative) or carrying only the Cre (Flox/Flox-negative). Males and females were used in the study. Induction of c-Myc knockout was performed between 4-6 weeks of age through daily intraperitoneal injections of 2 mg tamoxifen (#13258, Cayman Chemical, Ann Arbor, MI, USA) per animal for a total period of 5 days as previously described (Qi et al., 2022). Endothelial c-Myc overexpression mice were generated by crossing Tet-O-Myc (FVB/N-Tg(tetO-MYC)36aBop/J, Stanford University, Stanford, CA, USA) and Cdh5-tTA (FVB-Tg(Cdh5-tTA)D5Lbjn/J, Strain #013585 Jackson Laboratory, Bar Harbor, ME, USA) mouse lines in the presence of doxycycline diet (#TD.01306, Teklad, Indianapolis, IN, USA) to prevent the expression of the human c-Myc transgene. Overexpression controls consisted of littermates carrying the Tet-O-Myc genotype but lacking the transactivator (tTA-negative). Induction of c-Myc overexpression was performed between 4-6 weeks of age through withdrawal from the doxycycline diet. Only males were used in the study because the human c-Myc transgene is restricted to the Y-chromosome in this model. c-Myc knockout and overexpression in endothelial cells were confirmed by qPCR (Supplementary Figure S1).

For diet-induced obesity experiments, mice were exposed to control (#TD.08485) or high-fat diet (#TD.88137) (Teklad, Indianapolis, IN, USA) for a total period of 20 weeks.

Vasoactive response was assessed in mesenteric arteries harvested from control (CT) and endothelial c-Myc knockout (EC-Myc KO) mice using a pressure myograph system model 110p (Danish Myo Technology, Denmark) as previously described (Hernandez et al., 2019). Briefly, the mesenteric arcade was initially isolated and placed in physiological salt solution (PSS) for dissection of the second-order mesenteric arteries and preserve viability. Vessels were then mounted on two glass microcannulas and placed in the myograph chamber containing PSS solution at 37°C and under aeration with a special mix of 95% O₂ and 5% CO₂. Quality control was performed to confirm vessel viability prior to all measurements as previously described (Coats and Hillier, 1999). After confirmation of viability, arteries were washed with PSS, and pre-contracted with norepinephrine followed by assessment of endothelium-dependent vasodilation through exposure to increasing doses of acetylcholine every 5 min. Pressure-outside diameter curves were recorded at an isobaric condition of 60 mmHg as evidence of vasoreactivity response for different acetylcholine concentrations.

We performed longitudinal analysis of experimental animals maintained under a standard control or a western style high-fat diet. Body weight was collected once a week (high-fat diet) or once a month (aging), and animals were macroscopically evaluated for any visible abnormalities. Food consumption was monitored in EC-Myc KO animals. Any signs of abnormalities, cancer development and sudden deaths were recorded.

Whole-body energy metabolism was evaluated in CT and EC-Myc KO mice by indirect calorimetry using the Oxymax Comprehensive Lab Animal Monitoring System (CLAMS) (Columbus Instruments, Columbus, OH, USA) as previously described (Fonseca et al., 2014). Chambers were maintained at 22°C and O₂ levels were calibrated against a standard gas mixture prior to use. Mice were individually housed under light/dark cycles of 12-h and allowed to acclimate to the chamber for 48 h prior to data collection. Food and water were available ad libitum. After acclimation, O₂ consumption, CO₂ production, respiratory exchange ratio (RER) and heat production were collected at 26-min intervals for a total period of 48 h.

The redox state of skeletal muscle tissue harvested from CT and EC-Myc KO mice was estimated using a 3D fluorescence cryo-imaging system custom designed by the Biophotonics Laboratory at Florida Atlantic University (Ceyhan et al., 2023). Briefly, frozen tissues were embedded a day before imaging in a black absorbent medium. For imaging, the frozen tissue block was mounted to the sample stage, where its temperature was maintained at cryogenic temperatures (−10°C) for a higher quantum yield of fluorescence while retaining markers of metabolic state. A motor-driven stage and microtome blade allowed tissue slicing at 30 µm thickness. Images were acquired using a CCD camera (Retiga R6, Teledyne Photometrics, Tucson, AZ) with alternating filter wheels for nicotinamide adenine dinucleotide (NADH), and oxidized flavin adenine dinucleotide (FAD). The tissue was excited with a mercury arc lamp (200W lamp, Oriel, Irvine, CA). The broad light passes through excitation filters 350 ± 40 nm (UV Pass Blacklite, HD Dichroic, Los Angeles, CA) for the NADH channel and 437 ± 10 nm (440QV21, Omega Optical, Brattleboro, VT) for the FAD channel. All components of image acquisition operate with an automated virtual interface LabVIEW software (2022, National Instruments). The images of each slice were stacked in the z-direction to generate 3D images. Variables such as light intensity, illumination pattern, and dark current noise were considered for image processing. The redox ratio (RR = NADH/FAD) was calculated by dividing the fluorescence values of NADH over FAD images voxel by voxel. Representative images of NADH and FAD intensity are provided in Supplementary Figure S2.

Mice were fasted for 6-h with continuous access to water before the glucose load (i.p. bolus of 1.5 mg/kg body weight). Glucose levels were determined from a small drop of blood collected from the tail using a commercially available glucometer (AlphaTRAK^®, Zoetis, United Kingdom). Samples were collected before and 15, 30, 60, 90 and 120 min after glucose administration.

At endpoint, blood and major organs were collected for pathology analysis from all experimental groups. Tissues were macroscopically examined for visible signs of disease and organ weight was recorded. Blood was collected by cardiac puncture from fasted (approximately 6-h) and non-fasted animals. Plasma and serum were separated from blood for analysis of insulin and leptin levels using commercially available ELISA kits (#EZRMI and #RAB0334, Sigma-Aldrich Inc., St Louis, MO) per manufacture instructions.

Endothelial cells were magnetically sorted from hearts harvested from CT and EC-Myc OE mice based on CD31 expression using commercially available kits and instrument (Miltenyi Biotec Inc., Gaithersburg, MD). Briefly, harvested hearts were minced and dissociated with a mixture of enzymes (Multi Tissue Dissociation Kit 2, #130-110-203) using the gentleMACS Octo Dissociator system. After dissociation, cell suspension was cleared of debris (Debris Removal Solution, #130-109-398) and incubated with CD45 microbeads (#130-052-301) to exclude inflammatory cells, which may also express CD31. The CD45 depleted cell suspension was then incubated with CD31 microbeads (#130-097-418) for final sorting of endothelial cells. All steps followed manufacturer’s instructions for each specific kit used. RNA was extracted from sorted endothelial cells for analysis of mouse and human c-Myc by qPCR using Taqman probes (#Mm00487804_m1 and #Hs99999003_m1, respectively) per manufacture instructions (Life Technologies Corp, Carlsbad, CA).

RNA was extracted from soleus and gastrocnemius muscle harvested from CT and EC-Myc OE mice using TRI-reagent (#TR118, Molecular Resource Center Inc., Cincinnati, OH) per manufacturer’s instructions. RNA concentration was determined using a NanoDrop spectrophotometer (Life Technologies Corp, Grand Island, NY) and samples were outsourced for library preparation, sequencing, and bioinformatics analysis (Novogene Corporation, Inc. Sacramento, CA). Prior to library preparation, RNA integrity was evaluated using an Agilent BioAnalyzer 2100 (Agilent Technologies), and messenger RNA was purified from total RNA using poly-T oligo-attached magnetic beads. The library was checked with Qubit and real-time PCR for quantification and bioanalyzer for size distribution detection.

We used SigmaPlot (Inpixon) and Prism (GraphPad) Software for all graphs and statistical analysis. Sample numbers are indicated in all figure legends and p-values <0.05 were considered significant. For comparison between two groups, we performed Student’s t-test and considered two-tailed p-values. When samples did not pass the normality and variance tests, Welch’s t-test was used as an alternative. For comparison between four experimental groups, we performed two-way ANOVA. Holm-Sidak test was used for multiple comparison analysis. Tukey HSD tests were performed as a post hoc test to identify significant differences between individual groups. Two-way RM ANOVA was used for the interaction effect of the genotype with one continuous dependent variable. Individual animals are represented by dots in all graphs. Bars indicate the standard deviation unless otherwise stated.

Endothelial dysfunction is an early predictor of multiple pathological conditions. Little is known regarding early mechanisms that trigger endothelial dysfunction. Previous studies from our group have shown that the transcription factor c-Myc plays an important role in endothelial homeostasis and regulation of inflammation (Florea et al., 2013; Qi et al., 2022). These findings propelled us to investigate if c-Myc contributes to endothelial dysfunction ultimately impacting overall animal health. To confirm the association between c-Myc deficiency and endothelial dysfunction, we performed vasoactive response studies as impaired vasorelaxation is one of the first signs of dysfunctional endothelium. Mesenteric arteries of control (CT) and endothelial c-Myc knockout (EC-Myc KO) mice were harvested and tested for acetylcholine-induced vasorelaxation after constriction with norepinephrin. Our results show that loss of endothelial c-Myc significantly impacted vasorelaxation in response to acetylcholine relative to control. At the highest dose of acetylcholine tested (10⁻⁵ M), the ability of EC-Myc KO mesenteric arteries to relax was reduced by 29% relative to CT mice (58.2% ± 12.5% vs. 81.9% ± 5.6%, p = 0.031) (Figure 1).

We performed longitudinal analysis of CT and EC-Myc KO mice over a period of 16 months and found an expected increase in body weight over time, although findings were more pronounced in knockout animals relative to control (Figure 2A). At endpoint, body weight was significantly increased in EC-Myc KO females (31.67 ± 1.35 vs. 27.13 ± 0.61 g, p = 0.004) and males (44.2 ± 2.30 vs. 38.0 ± 1.03 g, p = 0.01) relative to CT (Figure 2B). When compared side by side, EC-Myc KO mice looked bigger in size than CT (Figure 2C).

Analysis of white adipose tissue (WAT) at endpoint in EC-Myc KO mice showed a significant increase by 76% in females (1.52 ± 0.20 vs. 0.86 ± 0.12, p = 0.007) and a trend increase by approximately 22% in males (2.74 ± 0.29 vs. 2.24 ± 0.22 g, not significant) (Figure 3A). Based on the observed increase in white adipose tissue (WAT) accumulation, in endothelial c-Myc deficient animals, we analyzed the circulating levels of leptin, which is expected to correlate with changes in adipocyte mass (Kiernan and MacIver, 2020). Our results showed a significant increase of approximately 47.98% in serum leptin in EC-Myc KO animals relative to CT (163.72 ± 21.48 vs. 110.64 ± 13.17 pg/mL) (Figure 3B). Representative images of fat deposits from CT and EC-Myc KO mice are shown in Figure 3C. Morphometric analysis of WAT from male and female EC-Myc KO animals revealed some sex-specific differences relative to CT (Supplementary Table S1). In EC-Myc KO females, we found a significant increase of 54% in the estimated number of adipocytes relative to CT (5.93 ± 0.75 vs. 3.85 ± 0.40, p = 0.025), but no changes in other parameters (Figure 3D). In males, the only significant difference between CT and EC-Myc KO was in the frequency of adipocyte size. EC-Myc KO males showed a higher percentage of very large-sized adipocytes relative to small sizes, while no differences were found in CT (Figure 3E).

Our observations described above suggest that loss of endothelial c-Myc causes an imbalance in energy metabolism. We analyzed food intake in CT and EC-Myc KO mice prior to the onset of weight gain and did not find any significant differences (data not shown). In the absence of alterations in calorie intake, one possible explanation for the observed increase in body weight in EC-Myc KO mice is a decrease in energy expenditure. Accordingly, we performed a series of metabolic studies with CT and EC-Myc KO mice by indirect calorimetry and found interesting sex-related differences between experimental groups. Similar results were found in dark and light periods (Supplementary Figure S3). A summary of all metabolic parameters is presented in Supplementary Table S2. In males, we found a significant increase in respiratory exchange ratio in EC-Myc KO relative to CT (0.91 ± 0.01 vs. 0.86 ± 0.02 RER, p = 0.015), without significant changes in other metabolic parameters (Figure 4D). However, in females, we found a significant decrease in heat production (0.39 ± 0.01 vs. 0.44 ± 0.03 kcal/kg/h, dark period p = 0.04) (Figure 4A), which was accompanied by a decrease in VO₂ (3203 ± 61 vs. 3687 ± 187 mL/kg/h, p = 0.007) (Figure 4B) and CO₂ (2791 ± 62 vs. 3241 ± 199 mL/kg/h, p = 0.01) (Figure 4C). No significant differences in RER were found in females (Figure 4D).

To further explore the impact of endothelial c-Myc loss in energy metabolism, we performed analysis of redox ratio in skeletal muscle harvested from CT and EC-Myc KO animals. Our results indicated a significant reduction of 51% in the mitochondrial redox ratio in EC-Myc KO relative to CT animals (1.53 ± 0.17 vs. 2.32 ± 0.2), which suggests a decrease in cellular energy production (Figure 5).

Our results described above indicate an important role for endothelial c-Myc in the maintenance of metabolic homeostasis. As such, we hypothesized that overexpression of c-Myc in endothelial cells would protect animals from developing overweight and obesity, as well as associated complications such as glucose intolerance and cardiovascular disease. To test this hypothesis, we performed a series of experiments in which we challenged control (CT) and endothelial c-Myc overexpression (EC-Myc OE) mice with a western-style high-fat diet (WD) over a period of 20 weeks. Exposure to WD promoted a gradual increase in body weight in both CT and EC-Myc OE mice (Figure 6A). This effect was significantly attenuated around 5 weeks post-exposure in EC-Myc OE relative to CT (31.10 ± 0.55 vs. 33.03 ± 0.83 g, p = 0.035) (Figure 6B). At 10 weeks post-exposure, the difference in body weight between both groups was lost. However, analysis of fat deposits revealed significant attenuation by 22% in visceral adipose tissue accumulation in EC-Myc OE mice relative to CT (408 ± 26.3 vs. 522 ± 43.9 mg, p = 0.028) (Figure 6C). No significant differences were observed in epidydimal and brown adipose tissue.

Quantification of systemic leptin levels showed a significant increase in CT animals under WD exposure relative to the control diet (48.98 ± 7.85 vs. 5.39 ± 2.12 pg/mL, p < 0.001), while the response was prevented in EC-Myc OE mice (12.46 ± 3.13 vs. 2.47 ± 0.23 pg/mL, not significant). Our findings revealed significant differences between EC-Myc OE and CT under WD exposure (12.46 ± 3.13 vs. 48.98 ± 7.85 pg/mL, p < 0.001) (Figure 6D).

One of the major complications associated with obesity is the development of insulin resistance. CT and EC-Myc OE mice have similar fasting glucose levels under normal diet conditions. Exposure to WD for 5 weeks was sufficient to significantly raise basal glucose levels in both CT (164 ± 8 vs. 129 ± 13 mg/dL, p = 0.013) and EC-Myc OE (147 ± 8 vs. 121 ± 7 mg/dL, p = 0.048) mice relative to the control diet. However, after 10-weeks of exposure to WD, although basal glucose further increased in CT animals relative to control diet (184 ± 15 vs. 130 ± 5 mg/dL, p = 0.001), it remained the same in EC-Myc OE mice (148 ± 6 vs. 113 ± 8 mg/dL, p = 0.02). Our findings indicated significant differences in basal glucose between EC-Myc OE and CT mice in response to long-term exposure to WD (148 ± 6 vs. 184 ± 15 mg/dL, p = 0.014) (Figure 7A).

We performed a glucose tolerance test (GTT) in all experimental groups to account for the development of glucose intolerance. Under the WD diet, the time to glucose peak in CT was longer than in EC-Myc OE (60 vs. 30 min). After 120 min, the level of glucose in EC-Myc OE mice under WD was almost completely back to baseline (211.00 ± 92.25 vs. 161.11 ± 43.44 mg/dL, 30% above baseline), while in CT animals, it remained significantly elevated (402.88 ± 121.50 vs. 183.83 ± 42.64 mg/dL, 120% above baseline) (Figure 7B).

At the 10-weeks endpoint, we measured systemic insulin levels and found a significant increase in CT animals exposed to WD relative to control diet (2.51 ± 2.18 vs. 0.44 ± 0.26 ng/mL, p = 0.001), which was significantly attenuated in EC-Myc OE animals (1.12 ± 0.38 vs. 0.48 ± 0.18 ng/mL, not significant). Our findings indicated significant differences in insulin levels between EC-Myc OE and CT mice in response to WD (1.12 ± 0.13 vs. 2.51 ± 0.77 ng/mL, p = 0.019) (Figure 7C).

The skeletal muscle plays an important role in energy metabolism (Mengeste et al., 2021). Based on our findings suggesting that endothelial c-Myc overexpression attenuates visceral fat accumulation and insulin resistance, we performed transcriptome analysis of skeletal muscle harvested from CT and EC-Myc OE mice. Exposure to WD for 10 weeks had a significant impact on the gene expression profile of both CT (1016 genes altered >1.5-fold) and EC-Myc OE mice (666 genes altered >1.5-fold) relative to control diet. Venn diagram analysis showed that both experimental groups shared common targets altered by diet exposure (243 genes) and that each experimental group had their own exclusive list of altered genes. It was noticeable that the number of genes in CT was almost double of what was found in EC-Myc OE (709 vs. 422 genes) (Supplementary Figure S4).

Comparison of the transcriptome profiles of WD-treated groups showed a total of 207 genes (128-up and 79-down) significantly altered >1.5-fold in EC-Myc OE (OE-WD) relative to CT (CT-WD). Among the most significantly altered targets, we identified Zbtb16 (2.35-fold, p = 1.39 × 10⁻¹⁹, Acsl3 (1.56-fold, p = 2.4 × 10⁻¹⁰) and Rabif (2.04-fold, p = 2.36 × 10⁻⁰⁷) as the top upregulated genes, and Rrad (−1.57-fold, p = 1.61 × 10⁻⁵) as the top downregulated. The top canonical pathways identified by Ingenuity software analysis that are affected by endothelial c-Myc expression are shown in Supplementary Figure S5. The S100 Family Signaling was among the top 5, comprising up- and downregulated genes. Some of the targets in this pathway include S100a3 (6.9-fold, p = 0.009, S100a14 (2.6-fold, p = 0.043) and Wnt10a (3.96-fold, p = 0.002).

We next analyzed the transcriptome profile associated with the exclusive response of EC-Myc OE and CT to WD relative to control diet (CTD). Pathway Analysis of upregulated genes differentially expressed showed significant increase in targets associated with extracellular matrix organization, collagen metabolism and organization for both groups. Analysis of downregulated genes revealed interesting differences between groups, including pathways associated with metabolism (Supplementary Figure S6). We focused our analysis on functions specifically related to metabolic disease based on our physiological and pathological findings. Both experimental groups showed changes in the expression of genes associated with diabetes, obesity, insulin resistance/sensitivity, glucose metabolism disorders, weight gain and energy homeostasis. However, the CT group showed a much higher number of genes altered under these function categories than EC-Myc OE (Figure 8). Interestingly, our analysis revealed other metabolic functions for EC-Myc OE that could account for the beneficial effects we observed. Some of the genes differentially expressed EC-Myc OE muscle have been related to weight loss, energy expenditure and glucose tolerance. Among the genes under these exclusive categories, we identified Socs3 (−1.83-fold, p = 0.009) as a common target in multiple pathways.

Obesity is a major risk factor associated with the development of cardiovascular disease. Based on the protective effect of endothelial c-Myc overexpression described above, we performed cardiovascular assessment of animals exposed to control and WD for 18 weeks by echocardiography. It is noticeable from looking at our data that endothelial c-Myc overexpression prevents several functional and structural diet-induced alterations observed in controls. CT animals showed an increase in myocardial performance index (MPI) relative to those fed control diet (0.54 ± 0.08 vs. 0.39 ± 0.04, p = 0.008), which was related to an increase in isovolumetric contraction time (IVCT) (10.56 ± 1.73 vs. 6.25 ± 1.10 m, p = 0.007). No significant changes in isovolumetric contraction time (IVRT) were observed. In addition, CT mice showed an increase in E/A ratio (3.99 ± 1.15 vs. 1.31 ± 0.13, p = 0.02). Interestingly, no significant changes in functional parameters were found in EC-Myc OE mice exposed to WD. At structural level, WD induced a significant increase in CT left ventricular mass (154.71 ± 12.50 vs. 114.12 ± 13.07 mg, p < 0.001), whereas no changes were found in EC-Myc OE (Figure 9). However, significant changes in wall thickness were found in both CT and EC-Myc OE mice after exposure to WD. We observed that EC-Myc OE mice under normal diet showed some baseline changes relative to CT, such as an increased fractional shortening (41.82% ± 1.78% vs. 32.64% ± 4.16%, p = 0.008), a reduced systolic diameter (1.94 ± 0.15 vs. 2.56 ± 0.39, p = 0.025) and increased thickness of the left ventricle posterior wall during systole (1.73 ± 0.18 vs. 1.30 ± 0.28 mm, p = 0.014) and diastole (1.33 ± 0.12 vs. 0.91 ± 0.25 mm, p = 0.008). A summary of all cardiac parameters is presented in Supplementary Table S3).